Recent advances in high-resolution imaging have permitted the development of new tools, notably micro- magnetic resonance (5MR) imaging and high-resolution peripheral quantitative computed tomography (HR- pQCT), that promise to provide a better profile of overall bone strength beyond areal bone mineral density (aBMD) by dual-energy x-ray absorptiometry (DXA). In this application, we seek to determine whether image- based microstructural and 5FE analyses can distinguish between individuals who have vertebral fractures from their counterparts without vertebral fractures. In the proposed project we advance the following hypotheses: 1. Morphological measurements and 5FE predictions of stiffness and failure load of the distal tibia and radius from ex vivo 5MRI and HR-pQCT correlate highly with those from 5CT and direct mechanical testing, and, furthermore, that parameters from the two peripheral sites parallel those in the vertebrae. 2. 5FE-derived estimates of elastic stiffness and failure load from in vivo 5MR and HR-pQCT images can differentiate between individuals with vertebral fractures from those without vertebral fractures better than microstructural measures derived by the two imaging modalities alone or aBMD by DXA. We plan to address the above hypotheses with the following specific aims: Specific Aim 1a: Perform 5MRI and HR-pQCT scans of the distal tibia and radius ex vivo under signal-to-noise and resolution conditions achievable in vivo, and compare trabecular and cortical bone microstructural measurements obtained in this manner with those from high-resolution 5CT. Specific Aim 1b: Compare the stiffness and failure load of whole bone segments of the distal tibia and radius as predicted by HR-pQCT and 5MR image-based nonlinear 5FE analyses to those predicted by 5CT image- based 5FE analysis and direct mechanical testing. Specific Aim 2a: Perform 5CT scans of lumbar vertebrae from the same subjects as the distal tibia and radius used in Aims 1a and 1b Specific Aim 2b: Compare trabecular and cortical bone microstructural measurements and 5FE predictions based on the imaging data obtained by HR-pQCT and 5MRI in Aims 1a and 1b with the 5CT measurements and direct mechanical testing of the corresponding vertebrae in Aim 2a. Specific Aim 3a: Apply the microstructural and 5FE techniques validated in Aims 1 and 2 to in vivo 5MRI and HR-pQCT scans from healthy women and compare these measurements between the two imaging modalities. Specific Aim 3b: Apply the microstructural and 5FE techniques validated in Aims 1 and 2 to the two peripheral imaging modalities and determine the effectiveness of methods in distinguishing between vertebral fracture subjects and their non-fractured peers using data from two imaging studies previously performed or currently in progress in the investigators' laboratories.